Amazon SageMaker Model Deployment with Lambda & API Gateway

Name: Amazon SageMaker Model Deployment with Lambda & API Gateway
Rating: 5.0 (11 reviews)

Amazon SageMaker • Amazon Lambda Amazon • API Gateway

Highest Rated

Created byJoe Cline

Last updated 1/2026

English

English [Auto],

What you'll learn

Create and train a SageMaker ML Model with XGBoost
Deploy the SageMaker Model to a model endpoint
Create a Lambda Function to invoke the model endpoint
Create an API Gateway app to invoke the Lambda Function
Create a VPC endpoint
Launch and EC2 instance and connect to it via SSH
Run a Linux cURL command to make a PUT request to the API

Course content

4 sections • 20 lectures • 1h 16m total length

Course Introduction2:06
Learn to train a binary classification model using the UCI Bank Marketing dataset, deploy it to a SageMaker real-time endpoint, and expose predictions via Lambda and API Gateway REST API.

Module 1: Introduction1:56
Explore how to deploy an AWS SageMaker model endpoint using Lambda and API Gateway, training and hosting the model, wrapping it in inference code, and exposing a secure REST endpoint.
Module 1: SageMaker Real-Time Endpoints Overview3:42
Transform your trained SageMaker model into a real-time endpoint that serves low-latency predictions via a scalable REST API, with artifacts in S3 and Docker inference code.
Module 1: AWS API Gateway + Lambda4:49
Discover how AWS API gateway and lambda create scalable, serverless apps that handle http and WebSocket requests with built-in security and monitoring.
Module 1: The UCI Bank Marketing Dataset5:56
Explore the UCI Bank Marketing Dataset to predict term deposit subscriptions using a supervised classification on Y, with personal, campaign, and macroeconomic features.
Module 1: Build, Train and Deploy a SageMaker ML Model Demonstration Part 17:02
Build and train an XGBoost model, split data 80/20, achieve about 89% accuracy, create model artifacts, and upload them to an S3 bucket using SageMaker JupyterLab.
Module 1: Build, Train and Deploy a SageMaker ML Model Demonstration Part 25:15
Load and prepare the UCI bank data, convert to DMatrix, load the XGBoost model from S3, deploy to a SageMaker endpoint, and test real-time predictions.
Module 1: Summary0:36
Learn to create and deploy a SageMaker endpoint with Lambda and API Gateway, save artifacts to S3, train with XGBoost, load data in JupyterLab, test the endpoint.
Module 1 Quiz

Module 2: AWS Lambda Introduction1:20
Wire up a SageMaker model endpoint to an AWS lambda function for HTTP predictions, grant least privilege, craft CSV payloads, parse responses, and configure a VPC endpoint.
Module 2: AWS Lambda: Run Code Without Servers3:27
Explore aws lambda, a serverless compute service that runs code and scales automatically. Write functions in python, node.js, java, or runtimes, triggered by events.
Module 2: AWS Lambda Prerequisites3:40
Build an AWS Lambda to accept test events, convert input to CSV format for XGBoost, and invoke a SageMaker real-time endpoint to return a yes-no JSON prediction via VPC endpoint.
Module 2: AWS Lambda Function Demonstration Part 11:12
Lock down VPC endpoint permissions by updating the policy with your SageMaker model endpoint ARN and AWS account number, then save and prepare to create a new Lambda function.
Module 2: AWS Lambda Function Demonstration Part 27:27
Create a secured AWS Lambda function to invoke a SageMaker endpoint from a VPC, using Python and boto3, with test events and IAM permissions.
Module 2: Module Summary0:45
Track progress in module 2 as you create a Lambda function, set up networking with SageMaker, grant SageMaker.invokeEndpoint permissions, test with a test event, then plan API Gateway security.
Module 2 Quiz

Module 3: Introduction3:43
Explore Amazon API Gateway as the secure, scalable front door for apps, translating HTTP or WebSocket requests, routing to Lambda or SageMaker, and handling auth, scaling, and monitoring.
Module 3: Amazon API Gateway Demonstration Part 16:54
Create a private REST API in API Gateway. Define a PREDICT resource, a PUT method with Lambda integration, require an API key via xAPIKey, and deploy to a Demo stage.
Module 3: Amazon API Gateway Demonstration Part 23:39
Create a source VPC policy from the template, replace with the API JSON, deploy the API, test the payload, and configure a usage plan and API key.
Module 3: End to End Testing from an EC2 Instance Demonstration Part 15:47
Launch an EC2 instance in the same VPC to perform an end-to-end test of the SageMaker model via the API, configuring a keypair, security group, and IAM role.
Module 3: End to End Testing from an EC2 Instance Demonstration Part 26:04
Learn end-to-end testing from an EC2 instance by setting up SSH access, adjusting security group inbound rules, securing the PEM file, and invoking a SageMaker API via VPC endpoints.
Module 3: Summary0:51
Explore deploying a SageMaker model via Lambda and API gateway, securing access with API keys, testing connectivity, and using VPC endpoints and curl-driven HTTP requests.
Module 3 Quiz

Requirements

Some Python programming is helpful but not necessary. Some basic API knowledge is helpful but not necessary

Description

Disclaimer: This course contains the use of artificial intelligence.

Due to a disability, I have used AI to assist with the production and editing of the audio in this course. The preview/promo video uses an AI generator of my image – but it is my image. I also use a "text-to-speech" AI tool to create AI audio of my voice recorded from an earlier date.

Oh, and I used it to create the image used in Udemy’s search results and in a couple of slides.

The ideas, most of the slides, script, demos, and the downloadable materials are of my own production. I hope this does not dissuade you from this course.

Now, back to the course description.

~ Joe Cline

Do you work in an AWS cloud environment and want to level up your skills by deploying a real machine learning model into production? Have you trained models before but felt unsure how to turn them into something users or applications can actually call?

If that sounds familiar, you’re in the right place.

Welcome to Amazon SageMaker Model Deployment with Lambda and API Gateway—a practical, hands-on course designed to help you connect the dots between machine learning and real-world AWS services.

In this course, you won’t just train a model and stop there. You’ll learn the complete deployment workflow, starting with building and training a machine learning model in Amazon SageMaker, deploying it to a managed SageMaker endpoint, and then exposing that model through a fully functional HTTP prediction API using AWS Lambda and Amazon API Gateway. By the end, you’ll understand how modern ML-powered applications are actually built in the cloud.

To keep things realistic, we’ll work with real data from the UCI Bank Marketing Database, and you’ll have access to the full data dictionary through the course downloads. Using this dataset, we’ll explore and answer a meaningful business question:

“Which bank customers are most likely to purchase a term deposit account?”

Every step of the process is demonstrated clearly, with all code provided in a downloadable JupyterLab notebook. You can follow along line by line, experiment freely, and reuse the project as a starter template for your own machine learning deployments.

Whether you’re a cloud engineer, data scientist, or ML practitioner looking to make your models production-ready, this course gives you practical, reusable skills you can apply immediately. No fluff—just clear explanations, real AWS services, and hands-on experience that will boost your confidence and your career.

Let’s turn your machine learning models into live, scalable APIs that actually deliver value!

Who this course is for:

AWS platform engineers, Data Engineers, Data Scientist, ML Engineers and Application developers

Amazon SageMaker Model Deployment with Lambda & API Gateway

What you'll learn

Explore related topics

Course content

Amazon SageMaker ML Model Deployment with AWS's Lambda and API Gateway1 lecture • 2min

Module 1: Creating an AWS Lambda Function7 lectures • 29min

Module 2: The Lambda Function6 lectures • 18min

Module 3: The API Gateway6 lectures • 27min

Requirements

Description

Who this course is for: